Why and how the same model for doc_embeddings and word_embeddings?

Nobody has claimed this yet.

BERT-based, or any transformer-based models output contextualized embeddings, which is correctly used for document embeddings generation. But to get word_embeddings, the same model is used, moreover, the array passed is just a list of raw candidate words, with no context, how will the word_embeddings hold any semantic meaning in that case?

In the BaseEmbedder class functionality is given to add word_embedding model, however in the "embed" method there is no way to differentiate between a list of documents and a list of words.

Start by reading the BaseEmbedder class and its embed method to understand how document and candidate-word inputs are currently handled. Clarify the expected distinction between contextual document embeddings and word embeddings, then establish the desired behavior and tests before attempting a change; the issue does not currently define a concrete completion condition.

Written by the indexing model from the issue text.